Yejin Choi
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
Safety Mid-Training: Internalizing LLM Safety as a Foundational Capability
According to the abstract, Zhengyue Zhao, Liwei Jiang, Yejin Choi, Chaowei Xiao and colleagues introduce safety mid-training, a stage between pre-training and post-training that develops safety and general reasoning together. They report it cuts average attack success rate by 72% to 81% on three Qwen3 base models while improving general capability.